Serhiy Storchaka added the comment: The condition that is determined by the PyObject_RichCompareBool() check is not that the element is already present, but that the element was present. Even if we will agree that such behavior change is appropriate, the inconsistency with dict makes it doubtful (originally set was implemented with dict).
I have no objections against tweaking so->table caching. I even think this part of the patch improves the code. ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue24681> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
